Article View: comp.ai.genetic
Article #90Re: Simple problems to solve genetically
From: James Rice
Date: Sun, 10 Jan 1993 15:58
Date: Sun, 10 Jan 1993 15:58
21 lines
1301 bytes
1301 bytes
At the risk of sounding boring again, it seems worthwhile pointing out that the lack of need to come up with a magical encoded representation is one of the key features of GP. Although I haven't put any cycles into thinking about blackjack (indeed games in general have the problem that you either need a minimax player to play against and possibly an enormous number of fitness cases or you need to co-evolve players - either way it's computationally expensive), there are a bunch of problems like this that are just screaming out to be addressed by a method that finds a program to solve the problem, not a mysteriously encoded character string. Rice.
Technical Details
Message-ID:
<RICE.93Jan10155821@HPP.stanford.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!morrow.stanford.edu!sumex-aim!riceReferences:
<1993Jan9.192644.24693@fcom.cc.utah.edu> <1993Jan10.054202.9817@ils.nwu.edu> <TOMF.93Jan10093713@ai.gte.com> <1993Jan10.213859.24981@seas.gwu.edu>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!morrow.stanford.edu!sumex-aim!rice From: rice@sumex-aim.Stanford.EDU (James Rice) Newsgroups: comp.ai.genetic Subject: Re: Simple problems to solve genetically Date: 10 Jan 93 15:58:21 Organization: Knowledge Systems Lab, Stanford University Lines: 21 Message-ID: <RICE.93Jan10155821@HPP.stanford.edu> References: <1993Jan9.192644.24693@fcom.cc.utah.edu> <1993Jan10.054202.9817@ils.nwu.edu> <TOMF.93Jan10093713@ai.gte.com> <1993Jan10.213859.24981@seas.gwu.edu> NNTP-Posting-Host: hpp.stanford.edu In-reply-to: freytag@seas.gwu.edu's message of Sun, 10 Jan 1993 21:38:59 GMT